A regression discontinuity design was used to assess the impact of a remedial education program, comparing credit course grades of students previously enrolled in remedial programs with those not previously enrolled. Results showed the remedial programs were not improving performance beyond expectation based on high school grades. (MSE)

This article presents a method for disentangling the effects of an intervention from changes that are due to growth and regression to the mean. The method is illustrated in the context of evaluating the effect of the Special Supplemental Food Program for Women, Infants, and Children (WIC) on infant growth. (Author/LC)

When the precise basis of selection effect on correlation and regression equations is unknown but can be modeled by selection on a variable that is highly but not perfectly related to observed scores, the selection effects can lead to the commonly observed "overprediction" results in studies of predictive bias. (Author/PN)

Used a current attrition study to show the usefulness of discriminative analysis and a cross validation technique applied to student nonpersister questionnaire respondents and nonrespondents. Results of the techniques allowed delineation of several areas of sample under-representation and established the instability of the regression weights…

Discusses the problem of dichotomous-dependent variables in regression analysis in student attrition studies. Proposes a new method as an alternative to regression analysis. (Author/RC)

When a test is made to separate a group into two distinct subgroups, as well as to provide an estimate of the mean performance of one subgroup (or both of them), such estimates will be biased due to the unreliability of the test. Suggestions for corrective procedures are presented. (RL)

A structural model for projecting public school enrollment by grade, program, and district is outlined. Main points of consideration are illustrated by an application to the problem of forecasting enrollment in Florida. (Author/GK)

Describes a study of graduate students at West Virginia University that examined the effects of Internet instruction on the attitudes of learners toward a new innovation before, during, and after the instruction. Responses from the stages of concern instrument are discussed, and results of regressions are described in appendices. (Author/LRW)

Employs a fixed-cost quadratic function to estimate multiple-output cost functions for 94 private universities in Japan for 1991. Outputs were undergraduate teaching, graduate teaching, and research. Results indicate ray economies of scale and both global and product-specific economies of scope. Product-specific economies of scale are shown for…

Describes how to teach the coefficient of determination and correlation coefficient. Includes a worksheet for correlation and regression activities. (KHR)

Challenges conclusions about infants' cognitive processing from prior research in favor of the importance of perceptual processes and the effects of stimulus novelty and familiarization. Considers problems with the two-test habituation design of earlier studies and proposes a new methodology that eliminates those problems. Describes a study the…

Used computer simulations to examine effectiveness of different criteria for measuring infant visual habituation. Found that a criterion based on fitting a second-order polynomial regression function to looking-time data produced more accurate estimation of looking times and higher power for detecting novelty effects than did the traditional…

A method is presented that simultaneously estimates cluster membership and corresponding regression functions for a sample of observations or subjects. This methodology is presented with the simulated annealing-based algorithm. A set of Monte Carlo analyses is included to demonstrate the performance of the algorithm. (SLD)

Major issues related to index-of-fit conjoint analysis were addressed in this simulation study. Goals were to develop goodness-of-fit criteria for conjoint analysis; develop tests to determine the significance of conjoint analysis results; and calculate the power of the test of the null hypothesis of random data distribution. (SLD)

Canonical weights and structure correlations are used to construct low dimensional views of the relationships between two sets of variables. These views, in the form of biplots, display familiar statistics: correlations between pairs of variables, and regression coefficients. (SLD)